On August 11, 2026, the Colorado attorney general's Department of Law filed proposed rules ("Automated Decision-Making Technology and Conversational Artificial Intelligence Service") specifying how employers must handle artificial intelligence in employment decisions. The rules implement two laws signed in May 2026—(1) the ADMT Act and (2) the Chatbot Safety Act—both effective January 2027. These laws regulate employers' use of automated tools to make "consequential decisions," such as hiring or termination, affecting employees and job applicants. 

Under these new rules, employers would have to disclose that automated technology is being used, and after a negative outcome, explain the tool's purpose, its role, the role of any human reviewers, and the principal reasons for the negative outcome "with specificity." For example, an employer who cannot explain how the tool influenced the decision or used personal data would not be compliant. On request, employers would also have to name the categories of information used, such as credit score, health information, or criminal history, and identify every source by name, tracing aggregated data back to its original source and each intermediary. People could also obtain the specific data used, including any score, ranking or, inference about them, and request corrections, with the adverse outcome paused where possible. They could further request "meaningful human review," which employers must acknowledge within ten days and complete within forty-five, using a trained, independent reviewer empowered to override the decision.

For employers using AI in hiring, promotions, or terminations, these rules would sharply raise the compliance bar—starting with knowing your own tools well enough to explain them. Practical steps include confirming that contracts with assessment platforms, background check providers, and data aggregators give you access to data-source information, and identifying, training, and documenting the people who will conduct human reviews. Human review is required only "to the extent commercially reasonable" under a multifactor test, but it is presumed reasonable where an outcome causes a "severe and irreversible denial of a basic human need," such as a termination affecting someone's livelihood. Because the rules are still in draft, it is worth tracking the rulemaking while planning for requirements likely to survive.

ABOUT BAKER BOTTS L.L.P.
Baker Botts is an international law firm whose lawyers practice throughout a network of offices around the globe. Based on our experience and knowledge of our clients' industries, we are recognized as a leading firm in the energy, technology and life sciences sectors. Since 1840, we have provided creative and effective legal solutions for our clients while demonstrating an unrelenting commitment to excellence. For more information, please visit bakerbotts.com.